The New Deal will continue into the next Parliament according to Gordon Brown. In a tub-thumping speech to the Labour Party conference, the Chancellor said that Government's duty was "to make the New Deal a permanent deal".

This means there will be continued financial support for jobsearch, training and job creation after the windfall tax revenues are exhausted. Decisions will actually be taken next summer at the conclusion of the next Comprehensive Spending Review, which will plan the Government's spending priorities into the next Parliament. In practice, it also looks certain that the £3.9bn allocated for the 5 years to 2002 will be underspent. As these revenues are ringfenced to New Deal so inevitably some funding will be available beyond the election.

So, where will the money come from? Firstly the Chancellor is looking at a very healthy budget surplus building up by election time. Some guess that the public finances will £10bn in credit - compared with a grim £50bn deficit during the last Conservative administration. But Brown knows the Keynsian credo that - in good times - surpluses should amassed and borrowings reduced. Only in bad times should the Government borrow and spend to counter a cyclical downturn. Inevitably Brown's Cabinet colleagues will press him heavily to spend more public funds whilst his Party will be hostile towards generalised tax cuts. And critics within his party will continue to press the argument that Blair's Government is spending less as a percentage of GDP than Thatcher's.

But a more subtle source of funding is available. As Paul Bivand shows (on page 15 of Working Brief, issue 108 ) the Government is already spending £7¾ billion a year on unemployment costs. More significantly, it is missing out on £12.8 billion of revenue from the potential PAYE and National Insurance contributions of people who want to work but cannot find work.

So every unemployed person receiving benefits "costs" the Government £10,200 per year. And even those who are drawing no benefit at all represent a tax loss of nearly £6,000 per year.

This is a calculation we have done before. Borrowing an idea first developed in the early 1980s by Adrian Sinfield it was calculated year after year by David Taylor at the Unemployment Unit. Then, we used it as a tool to show how the last Government's indifference towards the unemployed was a waste of financial resources as well as a cause of human suffering.

Today this methodology helps us illustrate an important aspect of emerging Government policy. Effort to help long term unemployed people back into work has a pay-off that more than justifies the initial expenditure on welfare-to-work measures. But more importantly, it reinforces that point the these measures must lead to sustained and lasting employment. If former claimants stay in long-term employment, it even makes sense to "risk" the budget allocated to benefits by spending it on back-to-work measures.

However, cautious Treasury types still balk at the idea of using income maintenance funds as a source for back to work expenditure. But the average claimant will have paid back even the most intensive assistance (up to £15,000) after 3 years of continuous employment. If they stay in work for ten years, we calculate a net present value to the exchequer of almost £20,000 per person.

The "benefit transfer" concept will soon be tested in the 15 areas earmarked to run the Government's "fully-fledged" Employment Zones. Almost half the £112million budget for "Personal Job Accounts" represents money that would otherwise have been spent on benefits.

But the Treasury's fingerprints were all over the final bidding guidance for these Zones. Because £55million of the benefit budget is being put "at risk", they have constructed an elaborate pricing grid that is heavily geared towards crude job-entry payment by results. At least half the Zones will be delivered by non-Government organisations because the Government wants to "encourage dynamism and innovation and bring in fresh approaches". They are "looking for diversity" and expect approaches which "are client focused and which genuinely motivate participants". That is excellent.

But insisting that external organisations bid for at least half the EZs, the Treasury has craftily offset the risk. By the end of October, we will know exactly which voluntary and commercial organisations have eventually bid for the Zones but this burden of risk means that bidders may not be that numerous.

Clearly, the Treasury is not yet convinced that, by diverting the benefit budget into expenditure that helps gets an unemployed person off benefit, the long-term gains will outweigh the risk that some claimants do not eventually get a job.

We believe the key to successfully drawing down the "exchequer costs" revenues is by offering even more intensive back-to-work measures that are carefully tailored to the individual's needs. Inevitably these will be expensive. But, as the Government says of the private sector, we need a high investment, high return economy. So it is with welfare-to-work.